Symposium on Erin Pobjie’s Prohibited Force: Reflections on the Utility of Erin Pobjie’s ‘Type Theory’ Approach to the Meaning of ‘Use of Force’

[James A. Green is Professor of Public International Law at the University of the West of England, Bristol, UK ,co-rapporteur of the International Law Association’s Use of Force Committee and a former editor-in-chief of the Journal on the Use of Force and International Law. His most recent book is Collective Self-Defence in International Law (CUP, 2024).]

Pål Wrange once wrote that the meaning of prohibited ‘force’ was “the subject of controversy par excellence in international law.” 

The question of what exactly Article 2(4) UNC (and its customary international law equivalent) prohibits has been explored throughout the UN era, often via more focused sub-questions: does economic or political ‘force’ violate the prohibition?; is merely providing weapons enough?; does ‘cyber force’ count?; is there a ‘de minimis’ threshold below which acts don’t qualify?; etc, etc. Yet, for all the associated uncertainty surrounding it, as Erin Pobjie notes at the start of her excellent new book, Prohibited Force: The Meaning of ‘Use of Force’ in International Law, the meaning of ‘use of force’ has not received anywhere near the sort of holistic, systematic treatment as has, say, the notion of ‘armed attack’ (see, e.g., here, here, and here). 

For me, there is no question that Prohibited Force is the most detailed and sophisticated treatment of the meaning of ‘use of force’ conducted to date, and the ‘type theory’ approach that it advances, in particular, provides a contribution to the literature of considerable note. In this post, I aim to explore the utility of that theory a little. It has substantial benefits as an analytical tool, but I nonetheless think we need to be cautious about overstating how far it can really take us in practice, especially in relation to ‘borderline cases’. 

What is Type Theory?

‘Type theory’ in a broad sense has existed at least since the start of the twentieth century, with versions of it notably appearing across – and at the intersections of – mathematics, computer programming, and applied logic (see, e.g., here and here). In Prohibited Force, Pobjie applies a variation of the theory to the meaning of ‘use of force’ in international law. The core argument is, first, that a prohibited ‘use of force’ is not a ‘concept’, for which there is a checklist of elements that must all be present for any given action to qualify. This is demonstrated convincingly by establishing that there are various ‘anomalous’ uses/non-uses of force. That is, actions that emerge from the subsequent agreement/subsequent practice of states as accepted instances of a prohibited ‘use of force’ but that do not possess the usually expected elements for a use of force (most obviously, physical acts and/or physical effects), as well as actions that do possess such features but are demonstrably not accepted uses of force. Pobjie thus argues that ‘use of force’ is instead a ‘type’, with a ‘basket’ of common ‘elements’, none of which are, in themselves, necessary, but which “are identified and weighed up to determine whether the threshold of the definition of ‘use of force’ is met.” (p. 195)

The great strength of Pobjie’s deployment of type theory is that it simultaneously establishes methodological consistency – the same framework can be applied to all use of force ‘candidates’ without caveat or exception – while also allowing for huge flexibility in any given determination, so as to encompass the wide range of activities that have become accepted by states to be ‘uses of force’. The theory is undoubtedly an elegant way of approaching the troublesome question of what a prohibited use of force ‘is’. It is also an approach built on principled, and robustly constructed, legal foundations. Pobjie meticulously applies the secondary rules of international law, and of treaty interpretation in particular, in establishing her list of ‘elements’ to be included in the type theory ‘basket’. 

Applying Type Theory in Borderline Cases

In the final chapter of Prohibited Force, Pobjie considers various ‘categories’ of action (such as those listed in Article 3 of the 1974 Definition of Aggression) as well as some specific instances from state practice, to demonstrate the application of type theory. It is persuasively argued that the theory can help to explain the fact that, in the context of actions at sea, for example, the 1975 Mayaguez incident was generally characterised as a ‘use of force’, whereas the 1967 Red Crusader incident was not. Notably, though, this part of the chapter focuses on examples that fairly clearly are (or are not) uses of force: type theory’s specific utility here is therefore that can help us to understand why.

Later in the chapter, though, Pobjie goes further and seeks to “show the usefulness of type theory for analysing controversial or borderline incidents…” (p. 212), through the more detailed discussion of two case studies: 1) the 2018 attempted assassination of Sergei Skripal in Salisbury, UK, through the use of the military-grade nerve agent Novichok; 2) the use of a Direct-Ascent Anti-Satellite Missile Test (DA-ASAT) to destroy a state’s own satellite, creating debris that then damages/destroys another state’s satellite. It is here, in relation to such instances where significant doubt exists as to whether an action is a prohibited use of force (‘borderline cases’), that I think one needs to be cautious as to how we characterise the utility of type theory. 

After applying type theory to her case studies, Pobjie concludes that they “illustrate how type theory can be applied to particular incidents to determine whether they are an unlawful ‘use of force’…” (p. 228, emphasis added). However, I’m not sure that the theory is capable of being that determinative. The case studies can equally be read as demonstrating how difficult it is, even with the theory, to reach a clear conclusion. 

As Pobjie concedes, whether the ‘elements’ that she identifies are indeed in the ‘basket’ is somewhat open to debate, as is their scope. For example, personally, while I found most of the arguments about the relevant ‘elements’ in the book persuasive, I was not entirely convinced by the importance that Pobjie ascribes to ‘intent’ in her constructed reading of Article 2(4), at least in relation to coercive and/or hostile intent (a concern I have discussed previously, in a different ad bellum context). And, as it happens, the importance of hostile intent (or its absence) is prominently emphasised as a factor in Pobjie’s analysis of both case studies. 

More importantly, the appropriate ‘weighing up’ of (and relative interaction between) the various ‘elements in the basket’ is considerably more uncertain than are the elements themselves. As emerges from the case studies, if different elements “are more heavily weighted” (p. 226), the conclusion can skew. One might note, for example, that the ‘risk of physical effects’ is an ‘element’ (or, more accurately, a factor to be taken into account in the assessment of the element of ‘effects’) that Pobjie discusses for both case studies. She concludes that the Salisbury attack did not reach the use of force threshold, and the implications of the ‘risk of physical effects’ is downplayed. Whereas, in the discussion of the DA-ASAT case study, while Pobjie ultimately is of the view that such action, too, probably would not qualify as a use of force, she is more equivocal, and the ‘risk of harm’ caused by space debris is presented as at least moving the dial closer to the threshold. 

I wonder, though, that if one of other factors that Pobjie argues is to be considered in assessing ‘effects’ – let’s say, “directness between the act and its harmful effect” (p. 132; discussed at pp. 134-136) – was emphasised more, those determinations might easily change. The release of a military-grade nerve agent amongst a civilian population with its direct risk to life versus an action against a state’s own satellite that poses only an indirect risk to other state’s satellites might suggest that the Sailsbury attack is the more likely candidate of the two to qualify as a use of force. I am not saying that the Sailsbury attack was a use of force, necessarily, or that emphasising ‘directness’ is a more appropriate way to ‘weigh up’ the elements. I merely seek to illustrate the uncertainty in applying type theory. Put differently, in the context of borderline cases, there is perhaps not a huge difference in practice between a ‘basket’ of elements and a ‘menu’ of elements – from which states can select factors to (de)emphasise to reach the conclusion that an action is or is not a use of force (as suits that state at that time, for whatever reason).

A related question, of course, is the location of the all-important use of force ‘threshold’. Pobjie refers throughout the book to this threshold – to be reached (or not) by combining (some of) the elements in the basket – but she never really tries to grapple with where it lies. Not unreasonably, because it surely is a variable, context specific line. But that in turn means that one can argue that what type theory ultimately offers is a more nuanced version of the ‘I know it when I see it’ approach to identifying a prohibited use of force (see p. 231). While the theory provides significant contextualisation to the determination, ultimately where the threshold to be crossed lies, and therefore when it is crossed, remains in the eye of the beholder. And given that the beholders are states, one might be sceptical about whether Pobjie’s own levels of legal-interpretative integrity will be applied to the question in practice.

Conclusion: The Utility of Type Theory

All this probably reads as me being unnecessarily harsh, and even unfair. It is important to be clear that I don’t intend here to criticise Pobjie’s approach for being unable to achieve ‘certainty’ in relation to the meaning of prohibited ‘use of force’. Of course, she never makes any claim to be able to provide certainty; indeed, she explicitly notes “the open questions and challenges involved” in applying the theory to the case studies (p. 212). The uncertainties I have tried to explore in this post are a feature of a degree of inevitable incoherence in the actions of states, and, more generally, of the pitfalls of legal interpretation as a field of enquiry in general. I’m not suggesting some kind of ‘failure’ on the part of type theory; I’m in full agreement with Pobjie when she says that 

[e]ven if the particular elements, their relationship and their combined threshold are debated, at the very least, the benefit of type theory is that it provides a shared language and coherent framework for legal analysis and scholarly debate regarding the content of a prohibited ‘use of force’ between States under international law.

p. 231

Type theory does indeed give us an invaluable starting point for a legally principled discussion about whether or not an action is a prohibited use of force. The importance of that contribution should not be understated because until now we have not had one. And, despite what I’ve said, the author’s assertion that “[i]n particular, the type theory framework is useful in borderline and novel cases…” (p. 228) may be accurate too. The theory can be ‘useful’ – perhaps even particularly useful – in relation to borderline cases, because this is where most uncertainty exists, and thus where any analytical help is the most welcome. 

The point I’m trying to make is simply that we should be clear about how far type theory can take us. For the most part, Pobjie is suitably cautious, but occasionally as I read her book, I felt she got a little swept up in the elegance of the solution offered by type theory so as to overstate its utility for her inquiry. Perhaps as result of its roots being in mathematics – a field that at least claims that it can produce certain outcomes – it is, indeed, tempting to view type theory as some kind of ‘calculation’ that can provide us an ‘answer’ as to what is/is not a use of force. But, of course, that would be wholly unrealistic in the messy reality of international law. In this context, type theory must be understood as (and celebrated for being) what it is: a new tool to frame and contextualise this inquiry.

Ultimately, as someone who has worked in the ad bellum field for over 20 years, I found that I, too, was extremely excited about type theory’s potential as I was reading Prohibited Force. It is a major addition to our toolkit for understanding what is, after all, a cornerstone rule of international law. With a little reflection, though, I realised that I mustn’t let that excitement mean that I expect more from type theory than it ever reasonably can deliver.